The present invention relates to a system having one or more storage devices and an information processing apparatus connected with the storage devices through a network.
In place of a configuration having a storage device directly connected with a computer (often referred to as a “host” or a “host computer”), today, another configuration is prevailing in which one or more storage devices are connected with a plurality of hosts through a network. The connecting configuration of storage device(s) through a network is called a storage area network (referred to as a “SAN”). The SAN has been heretofore configured by using fibre channel technology (referred to as “FC”). Hereafter, a SAN configured by using the FC technology is called an FC-SAN.
Conventionally, in the FC-SAN, hosts and storage devices are divided into groups each called a zone through the use of a function of a switch called zoning in order to limit accessible hosts to the storage devices by managing the groups. For example, as a method of setting zones in FC-SAN, the setting method has been disclosed as shown in the Official Gazette of the Japanese Patent Laid-open No. 2003-141055.
On the other hand, today, a remark is focused on an IP-SAN which is a SAN configured by using an IP (Internet Protocol) network. Like the zoning of the FC-SAN, the IP-SAN enables system administrators to limit accessible host(s) to the storage devices through the use of a VLAN (Virtual Local Area Network) of a switch. The VLAN is a function of dividing one physical network into a plurality of logical networks. When one physical IP-SAN is divided into a plurality of logical IP-SANs and the hosts and the storage device to be used by the hosts are connected with each logical IP-SAN, a false access by an administrator of a certain host to a storage device to be used by another host may be prevented. Also, another false access by a cracker having intruded into a host to a storage device to be used by another host may be prevented, for reducing damage caused by cracking into a minimum. Further, if a certain host is infected with a virus, reinfection to another host may be prevented, which also leads to reducing damage caused by the virus to a minimum.
In a case that a host communicates with a storage device through the aforementioned IP-SAN, iSCSI is mainly used. iSCSI (Internet Small Computer System Interface) is a protocol in which SCSI protocol is encapsulated with TCP/IP (Transmission Control Protocol/Internet Protocol). Of devices and software for performing communications through iSCSI, software and devices that send commands for requesting writing or reading of data and that send data to be written are each called an iSCSI initiator. On the other hand, devices and software for receiving the write command and the data from the iSCSI initiator, writing the data onto a storage device, receiving the read command from the iSCSI initiator, reading data from the storage device, and transmitting the data to the iSCSI initiator are each called an iSCSI target. iSCSI initiators and iSCSI targets are collectively called iSCSI nodes. Each iSCSI node has an identifier called an iSCSI name.
When an iSCSI initiator logs in an iSCSI target, the iSCSI name and the IP address of the iSCSI target are required. The obtention of such information of the iSCSI target is called “discovery”. For each host operating as an iSCSI initiator, work of specifying the iSCSI name and the IP address of the iSCSI target is quite heavy in load. Hence, one or more methods of not setting the iSCSI target information into the iSCSI initiator but causing the iSCSI initiator to perform the discovery are regulated in the iSCSI protocol.
As one of those methods, a method of connecting a name management device with an IP-SAN may be referred. The name management device is a device that manages combinations of an iSCSI name and an IP address of each iSCSI node. As a communication protocol between the name management device and the iSCSI nodes, iSNSP (Internet Simple Naming Service Protocol), SLP (Service Location Protocol) and so forth may be used. Hereafter, an operating process about the discovery executed through the use of the name management device will be described below. At first, a storage device that operates as an iSCSI target, after started, registers its own iSCSI name and IP address with the name management device. On the other hand, a host that operates as an iSCSI initiator, after started, inquires the iSCSI name and the IP address of the iSCSI target in which the iSCSI initiator may log from the name management device and then obtain such information. As described above, since the name management device unifies management of the combinations of the iSCSI name and the IP address, the work of setting the information to the hosts may be greatly reduced.